Deep 6-DoF Tracking of Unknown Objects for Reactive Grasping

TL;DR

This paper introduces a real-time 6-DoF tracking method for unknown objects using Siamese Networks and ICP, enabling robots to perform reactive grasping without additional training, even in noisy or occluded conditions.

Contribution

The authors present a novel, training-free 6-DoF object tracking approach combining Siamese Networks with ICP for robust, real-time reactive grasping of unseen objects.

Findings

The method is robust to noise and occlusion.

It enables successful grasping of unseen objects.

The system operates without additional training.

Abstract

Robotic manipulation of unknown objects is an important field of research. Practical applications occur in many real-world settings where robots need to interact with an unknown environment. We tackle the problem of reactive grasping by proposing a method for unknown object tracking, grasp point sampling and dynamic trajectory planning. Our object tracking method combines Siamese Networks with an Iterative Closest Point approach for pointcloud registration into a method for 6-DoF unknown object tracking. The method does not require further training and is robust to noise and occlusion. We propose a robotic manipulation system, which is able to grasp a wide variety of formerly unseen objects and is robust against object perturbations and inferior grasping points.